RISK OF LYMPH NODE METASTASIS IN T2 COLON CANCER IN THE NETHERLANDS - A NATIONWIDE POPULATION-BASED COHORT STUDY: PUSHING THE BOUNDARIES OF LOCAL TREATMENT?

UEG Presentation

Log in to access this content.

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Log In Create a free account

Not sure what you can access? Learn more about account types.

Introduction

The implementation of the national bowel-screening program for colorectal cancer has had a significant impact on detecting early colorectal cancer (pT1 & pT2 carcinomas). In recent decades, there has been a shift towards minimally invasive approaches for treating T1 colon carcinomas (CC), favoring local resection techniques over major surgeries. It is unknown whether risk stratification could establish organ preservation in T2 CC.

Aims & Methods

We aimed to evaluate the incidence of lymph node metastases (LNMs) in patients with pT2 CC in the Netherlands and to develop a prediction model to estimate the risk of LNMs which could be helpful to compare the risk of LNMs against the age-specific adverse events associated with segmental surgery. A retrospective nationwide cohort study was performed including all consecutive patients diagnosed with pT2 CC who underwent (completion) surgery between 2012 and 2020 in the Netherlands. Patients were identified by use of the Dutch ColoRectal Audit (DCRA). Exclusion criteria were: (1) neoadjuvant therapy; (2) synchronous CRC; (3) hereditary predisposition (4) history of IBD; and (5) non-adenocarcinoma. Data obtained from the DCRA were linked to the Dutch Nationwide Pathology Databank (PALGA). A subgroup analysis was conducted on patients up to 70 years old, who were treated between 2018 and 2020, to examine the association between mismatch repair (MMR) status and the presence of LNMs. Predictive factors for LNMs were evaluated with univariate- and multivariate logistic regression. Four machine learning (ML) models were developed and validated to predict LNM, including generalized linear modeling (GLM), GLM-Least Absolute Shrinkage and Selection Operator (GLM-LASSO), random forest (RF), and gradient boosting (XGB).

Results

In total 9,803 patients were included. The overall 30-day complication and mortality rates were 25.2% and 2.2%, respectively. In elderly patients (age ≥ 74), the complication and mortality rates increased to 30%, and 3.5%, respectively. LNMs were found in 1877 out of 9803 patients (19.1%). Younger age (OR 0.98, 95% CI 0.979 ─ 0.990), left-sided CC (OR 1.5, 95% CI 1.4-1.7), poor differentiation (1.7, 95% CI 1.4-2.2), and lymphovascular invasion (LVI) (OR 4.1, 95% CI 3.6-4.7) were independent risk factors for LNMs. In the subgroup analysis, deficient MMR (dMMR) status was found to be associated with a significantly lower risk of LNMs (OR 0.3, 95% CI 0.2-0.5). GLM demonstrated the highest prediction accuracy, achieving AUCs of 0.67 for the full model and 0.68 for the model focusing on patients up to 70 years old. The risk of LNMs in the lower-risk group (age ≥ 74, right-sided CC, LVI absent, good/moderate tumor differentiation) was 10.7%. In the absence of other risk factors, patients with a dMMR carcinoma had a risk of LNMs of 6.1%.

Predicted probability (%) of LNM in patients with T2 colon cancer
		Right sided CC			Left sided CC			dMMR / pMMR
		Age ≥ 74	66 - < 74	Age < 66	Age ≥ 74	66 - < 74	Age < 66	< 71
Good/Moderate differentiation (G1-G2)	LVI -	10.7	12.6	14.1	15.6	18.2	20.2	6.1 / 17.2
Good/Moderate differentiation (G1-G2)	LVI +	33.2	37.5	40.6	43.4	48.1	51.4	19.3 / 43.1
Poor differentiation (G3)	LVI -	16.5	19.3	21.4	23.4	26.9	29.6	27.9 / 55.1
Poor differentiation (G3)	LVI +	45.2	49.9	53.2	56.1	60.6	63.7	58.6 / 81.8

Complication rate		30.1	22.6	18	30.1	22.6	18	19.2
Mortality rate		3.5	1	0.5	3.5	1	0.5	0.7

Conclusion

The predicted probability of LNMs should be balanced against the risks associated with segmental colon surgery. The findings of this study could serve as valuable guidance for clinicians, enabling them to inform patients about these competing risks before making a shared decision.

RISK OF LYMPH NODE METASTASIS IN T2 COLON CANCER IN THE NETHERLANDS - A NATIONWIDE POPULATION-BASED COHORT STUDY: PUSHING THE BOUNDARIES OF LOCAL TREATMENT?

Julia Hanevelt 1, Richard M. Brohet 1, Leon M.G. Moons 2, Frank P. Vleggaar 2, Erik Van Westreenen 1, Wouter De Vos Tot Nederveen Cappel 1

1 Isala, Zwolle, Netherlands

2 University Medical Center Utrecht, Utrecht, Netherlands

Event

UEG Week Vienna 2024

Topics

Colorectal Digestive Oncology

Submission format

Abstract

Session

Local treatment for locally advanced colon cancer: How to improve outcome

Citation

United European Gastroenterology Journal 2024; 12 (Supplement 8)

Published

2024

UEG Presentation

COMPUTER AIDED QUALITY CONTROL SYSTEM IMPROVES BARRETT’S NEOPLASIA DETECTION IN AN EX-VIVO SETTING

Log in to access this content.

Free for all myUEG account holders. Your access level is set automatically based on your occupation. Medical professionals get full access to all content. If you are a non-medical user, you can only access UEG Week content from congresses you attended.

Log In Create a free account

Not sure what you can access? Learn more about account types.

Introduction

Early detection of neoplasia in Barrett’s Esophagus (BE) remains challenging. Although computer-aided detection (CADe) systems are increasingly available to support endoscopists, successful CADe implementation largely depends on the quality of endoscopic images, which can vary significantly in clinical practice. This study introduces an innovative computer-aided quality (CAQ) system designed for BE endoscopy, capable of assessing various image quality parameters, providing real-time feedback to the endoscopist, and operating in conjunction with an established CADe system.

Aims & Methods

The CAQ system was trained on 4,926 images from 243 BE patients to assess both objective (e.g., blurriness, illumination, and specular reflections) and subjective (i.e., mucosal cleanliness and esophageal expansion) image quality factors. The system is able to provide real-time feedback on these quality parameters to the endoscopist. Additionally, it is configured to function as a preprocessing filter for a previously developed CADe system, automatically excluding lower-quality images. To validate the CAQ system’s performance as a standalone tool, we constructed an endoscopic image quality test set, which included 647 images from 51 non-dysplastic BE patients across eight hospitals. These images were annotated for objective (e.g. motion blur, lighting) and subjective (i.e. esophageal expansion, mucosal cleanliness) image quality parameters by two experts. Additionally, to evaluate the CAQ system’s role as a preprocessing step for an established CADe system, we created the Barrett CADe test set, containing 956 frames from 62 neoplastic BE patients and 557 frames from 35 non-dysplastic BE patients from 12 Barrett referral centers. Outcome measures included agreement scores with experts for standalone image quality assessment and sensitivity, specificity, and Area-under-the-Curve (AUC) for BE neoplasia detection.

Results

As a stand-alone system, the CAQ system achieved Cohen’s Kappa scores for agreement with expert annotators of 0.73, 0.91, and 0.89 for objective image quality parameters, mucosal cleanliness, and esophageal expansion, respectively—comparable to inter-annotator scores of 0.73, 0.93, and 0.83. When used as a preprocessing filter for the CADe system, the CAQ system improved the CADe's sensitivity from 82% to 90% and the AUC from 87% to 91%, while specificity remained stable at 75% (Table 1).

Table 1. Performance of the CADe system for Barrett’s neoplasia detection with and without the CAQ system.

Evaluated by	Sensitivity, %, (95% CI)	Specificity, %, (95% CI)	AUC, %, (95% CI)
CADe	82.2 (79.7 - 84.5)	74.7 (70.9 - 78.1)	87.2 (79.9 - 94.5)
CAQ + CADe	89.9 (87.2 - 92.1)	74.7 (70.0 - 78.9)	91.4 (82.0 - 100)

Conclusion

This study introduces the first CAQ system specifically aimed at automated quality control in BE endoscopy. As a standalone tool, the system effectively distinguishes poorly from well-visualized mucosa. When combined with an established CADe system, it enhances detection rates for neoplasia, highlighting its potential to improve diagnostic accuracy in BE surveillance.